In the previous funding period, we demonstrated that diabetic individuals and mice demonstrate dysfunctional endothelial progenitor cells. These bone marrow (BM) derived cells are released into the circulation in an abnormal manner due to diabetes-induced BM neuropathy. This BM neuropathy leads to dysregulation in the circadian release of these cells into the circulation. These BM changes precede development of diabetic retinopathy (DR) linking BM neuropathy for the first time to DR. Diabetic BM supernatants, but not control, contain high levels of monocyte/macrophage-colony stimulating factor (M-CSF) and diabetic mice exhibit an increase number of monocytes in the retina and circulation. Based on these findings, we hypothesize that: Diabetic BM denervation initiates HSC/progenitor dysfunction that leads to an increase in total monocytes generated, this, coupled with increased expression of CCL2 (monocyte chemoattractent protein-1) by retinal glia results in excessive levels of the pro-inflammatory (M1) monocytes in the diabetic retina. In addition, diminished levels of CX3CL1 (fractalkine) expression by dysfunctional neurons heightens the inflammatory response of resident microglia. This imbalance leads to a neural inflammation that accelerates DR pathogenesis. We will test this hypothesis both in vitro and in vivo through the simultaneous exploration of the following aims: Aim 1: To determine whether diabetes induced BM neuropathy is responsible for increased M-CSF production by BM stromal cells which shifts hematopoiesis towards generation of excessive numbers of monocytes, in particular pro-inflammatory Gr1+/CCR2+/CX3CR1lo monocytes, rather than the protective Gr1-/CCR2-/CX3CR1hi monocytes. Aim 2: To determine the roles of CCL2 and CX3CL1 in DR inflammation and to test whether restoration of the balance of CCR2+CX3CR1lo and CCR2-/CX3CR1hi monocytes and simultaneous restoration of levels of endothelial progenitors to retinal vasculature will prevent o reverse DR in T1D mice. Aim 3: Our hypothesis predicts that circulating pro- inflammatory Gr1+/CCR2+i/CX3CR1lo cells will cause endothelial cell activation, promote leukostasis and result in enhanced retinal permeability whereas Gr1-/CCR2-/CX3CR1hi cells will protect the BRB in diabetes. We will examine the effects of manipulating these subsets on the progression of DR. Targeting monocyte subtypes may represent an ideal strategy for DR prevention and treatment. CellMax?